Why do Poor students underachieve at A-level?

Table of Contents

16-19 year old students who are eligible for free school meals underachieve by 3 A-level grades compared to their wealthier peers.

This is primarily because of lower prior attainment at GCSE, but also because poorer students take fewer and different qualifications (BTECs more likely than A-levels).

This is according to some recent quantitative research published in 2021 by Tuckett al: Measuring the Attainment Gap in 16-19 Education (1).

The rest of this post summarises and evaluates this research.

Methodology

Sampling

The sample of students was about as close to a ‘total sample’ as you can get. It included all students at the end of their 16-19 study at a state-maintained school or college other than those on apprenticeship programmes.

Measuring Disadvantage

To measure (or ‘operationalise’) disadvantage the researchers used students’ free school meal status during their last six years of school (prior to key stage 4) as the indicator.

They also conducted some analysis using a measure of persistent disadvantage which was defined as any students who had been eligible for Free School Meals for 80% of the previous 6 years.

Measuring Educational Attainment

To measure educational attainment the researchers used the best three qualifications achieved by the end of 16-19 education.

Interestingly, they used two different weighting systems to take account of the different types of qualification students achieved results in: the main difference being between A-levels and BTEC subjects.

For on measure of attainment they treated all level 3 qualifications as being equal, giving the same weight to all level three courses with the same guided teaching hours – so all A-level subjects had the same ‘achievement’ rating as all level 3 BTEC courses. (This is the standard way of measuring Attainment used by UCAS).
They also used a second measure of attainment by adjusting the above for the economic value associated with the different qualifications. Thus science based A-levels would receive a higher score than BTEC business studies, because the kind of jobs students who achieve A-levels in physics, chemistry and biology go on to do are higher paid.

Analysis of results

This is a bit technical for A-level students, but they use Regression analysis. More specifically they used ordinary least regression squares holding attainment as the dependent variable with students clustered into institutions.

They also used Oaxaca Blinder decomposition to find out how much of the difference in achievement between disadvantaged students and non-disadvantaged students were down to a specific variable.

The rest of this post outlines the findings of this study.

How many 16-19 year old students are disadvantaged

in 2019 there were 119, 497 16-19 year old students who were classified as disadvantaged, meaning they had been eligible for free school meals for at least one of the previous six years.

119, 497 students is equivalent to almost 25% of of the total number of 16-19 students in 2019 which was 497, 541.

Year	Disadvantaged	Non-disadvantaged	Total students
2017	119. 980	385. 178	505, 158
2018	120, 049	378, 839	498, 888
2019	119, 497	378, 044	497, 541

How big is the attainment gap between ‘poor’ students and the rest?

By age 19 poor (disadvantaged) students are almost 3 A-level grades behind non disadvantaged students, if we give all A-levels and BTECs equal waiting.

It we weight different qualifications according to their economic value then poor (disadvantaged) students are more than 4 grades behind non disadvantaged students.

The disadvantage gap narrowed slightly between 2017 and 2019, but not significantly and more recent evidence suggests that the Pandemic increased this gap again.

interestingly in terms of ‘average’s it makes quite a difference whether you use the Mean score which they use here or the Median – there are significant numbers of 16-19s who don’t achieve, so by including those you drag the results of the ‘disadvantaged’ down because the extreme majority of those who get no results are disadvantaged!

Why do poor students get worse results?

Regression analysis shows that:

Prior attainment explains 39 per cent of the total gap,
the type of qualifications entered explains 33 per cent.
the average prior attainment of students’ peers explains 12 per cent

The researchers also noted that fourteen per cent of the disadvantage attainment gap cannot be explained by student or institution characteristics, equivalent to almost half an A level grade. This could be the continued effect of disadvantage itself, and/or it could be due to differences in unobserved characteristics such as health or motivation

Disadvantaged students take different qualifications

Disadvantaged students are more likely to take vocational and technical qualifications. They also tend to enter fewer, and lower level, qualifications.

Taken altogether these differences explain 33% of the attainment gap, mainly because fewer and lower level qualifications mean lower point scores at age 19!

The disadvantage gap and ethnicity

There are significant variations in the disadvantage gap by ethnicity.

Poor white students underachieve by around 4.5 A level grades compared to their richer peers, equivalent to almost an entire A level.

The disadvantage gap is smaller for all other ethnicity groups.

Policy suggestions

One specific policy suggestion is to extend the pupil premium to 16-19 year old students. This means that colleges should receive extra funding for each student they enrol who is eligible for free school meals and have to spend that money supporting disadvantaged students with extra lessons for example,

Strengths and Limitations of this study

This study is very useful because it fills a research gap focussing specifically on the post-16 education sector.

It shows that the disadvantage gap at GCSE level continues into post-16 education and that poor prior attainment explains most of the achievement gap in post-16 education. It also shows that qualification type explains a significant amount of the gap with poor students having to ddo fewer and lower level qualifications.

Sampling is very strong with a near total sample used.

This is also an example of a study which uses some innovate research methods – through the use of multiple measures. I especially like the measure which weights qualifications for future economic value because anyone who has worked in a sixth form environment knows that not all A-levels and BTECs are worth the same, even though UCAS insists on giving them equal weight.

In terms of weakness I don’t like the fact they do most of their analysis using the mean, I’d rather the median – I think it’s fairer to compare students who actually do qualifications!

One final limitation is the time-scale – published in 2021 but it’s only showing data up to 2019, and with the Pandemic, we are now in a different era so this is already in need of an update!